The present invention relates to web application testing and, more particularly, to detecting a broken point in a web application automatic test case.
For many web applications, automatic test scripts can be built to cover basic user scenarios. Whenever a developer modifies some code, e.g., when a new version of a product is launched, the scripts or their subsets need to be re-run to ensure that the functions of the product are still correct. However, in many circumstances, due to changes to user interfaces (UIs) of the web application, the test scripts themselves are no longer valid. That is to say, these automatic test scripts need to be updated according to the UI changes made by developers. Currently, UI changes made by the developers are not synchronized to the automatic test scripts, and thus, the testers need to repeatedly run the automatic test scripts, so as to find broken points without any hints, and then manually fix them. For a large project, determining all of the broken points in an automatic test case, as well as fixing them, can be inefficient and yield inaccurate results.